Mobile phone price fixed on the various factors like processor, RAM, Internal memory etc., There are 21 features given in the dataset. Classify the mobile price range as 0,1,2 or 3 based upon the features given in the dataset.

Class 0 – low cost

Class 1 – medium cost

Class 2 - high cost

Class 3 – very high cost

Use SVM algorithm for classification.

EDA

pc - Primary Camera mega pixels

fc - Front Camera mega pixels

sc_h - Screen Height of mobile in cm

sc_w - Screen Width of mobile in cm

m_dep - Mobile Depth in cm

px_width - Pixel Resolution Width

px_height - Pixel Resolution Height

ram - Random Access Memory in Mega Bytes

int_memory - Internal Memory in Giga Bytes

four_g - Has 4G or not

three_g - Has 3G or not

dual_sim - Has dual sim support or not

battery_power - Total energy a battery can store in one time measured in mAh

touch_screen - Has touch screen or not

clock_speed - speed at which microprocessor executes instructions

n_cores - Number of cores of processor

wifi - Has wifi or not

blue - Has bluetooth or not

mobile_wt - Weight of mobile phone

talk_time - longest time that a single battery charge will last when you are

price_range - This is the target variable with value of 0(low cost), 1(medium cost), 2(high cost) and 3(very high cost).

Model Building

After applying StandardScaler

After applying MinMaxScaler

Conclusion

The high accuracy of 97% in the first case indicates that the model is able to predict the mobile phone price range with high accuracy when features are used without any scaling.

The low accuracy of 26% in the second case shows that the model's performance drastically drops when features are not scaled properly. This implies that scaling is essential for improving the model's performance.

The accuracy of 96% in the third case suggests that using MinMaxScaler helps in improving the model's performance compared to the second case but is slightly lower than the first case. MinMaxScaler scales the features to a specific range, which seems to help the model make better predictions.

Overall, scaling the features using appropriate scalers can have a significant impact on the model's performance, and it is essential to choose the right scaling method depending on the data and the machine learning algorithm being used.